Several assumptions underlie the present proposal. First, family processes can lead to mental health problems for the child and caretaker, and these problems will manifest themselves at transition points in the child's development. Second, family processes will also account for the child's competencies in both social and cognitive areas and different competencies will result in varying self-evaluations. Third, families must be studied over time if we are ever to understand the processes by which outcomes are influenced. We are currently following 185 families with children from 18 months through the transition into school at age 7. We have found that parents who are unprepared to deal with their children's bids for autonomy during the toddler period have great difficulty; this difficulty persists until the age of 4. If the parents do not quickly develop a whole new set of skills, the children do not learn age-appropriate behaviors. As the children's behaviors become less age-appropriate, the parents' skills are less effective, decreasing everyone's ability to cope with the next crisis. Stressful events or marital discord occur often enough to precipitate long-term problems in the parent-child relationship. Disruption in management skills reduces the parents' ability to instruct or to provide guided learning. These same children have been followed through the major transition of school entry, and we have added another 100 families from high-risk areas. We are examining family processes through a combination of parent self-report measures, laboratory tasks, and home observations. With the older group of children, we will also examine the child's perception of family roles and functions. Parenting variables ill be assessed using multiple methods in three domains (behavior, affect, and cognition). We will look at child outcome variables in three areas (cognitive competence, prosocial behaviors, and problem behaviors). Since poor peer relations are a major predictor of mental health problems, it as important to gain additional information in this area of functioning. We added an examination of the child's peer relations at age 7 to the current study. We will continue collecting data in this area in the proposed study.